INDEX

Explanations

terms related to expansion, enlargement, and growth in a variety of contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

expansion and enlargement

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Einstellungen

-0.49

 AppCompatTheme

-0.49

 transfieras

-0.44

SequentialGroup

-0.44

 للمعارف

-0.43

Pyx

-0.43

PLEMENT

-0.42

 strictest

-0.42

ancourt

-0.41

ungsbedingungen

-0.41

POSITIVE LOGITS

 expansion

0.69

 enlarge

0.68

 enlargement

0.63

 expands

0.62

 size

0.60

 enlar

0.60

 enlarged

0.60

 expanded

0.58

 расши

0.58

 Enlarge

0.57

Activations Density 0.105%

terms related to expansion, enlargement, and growth in a variety of contexts

expansion and enlargement

No Comments

No Known Activations